Mobile telecommunications networks are usually arranged according to a cellular structure comprising a plurality of cells, each cell being defined as the set of elementary territory areas (also referred to as “pixels”) served by the radio-electric signal radiated from a respective Base Radio Station (BRS), or antenna.
Among the known cellular networks, networks using the CDMA or WCDMA technique have the peculiarity that a same frequency band (or “channel”) can be re-used in the various cells. Therefore, the passage of a mobile communications terminal from one cell to another, contiguous cell (an event called “handover”) can be managed by using the same frequency, according to a mechanism called “soft-handover”; this mechanism provides that, in particular geographic areas, called “soft-handover areas” or “macro-diversity areas”, the mobile communications terminal is able to decode signals from, and therefore to exchange information with many antennas and consequently with many BRSs.
The location of the macro-diversity areas and their dimensioning are highly important factors for the correct operation and dimensioning of the network cells' apparatuses: a mobile communications terminal operating in macro-diversity uses resources of all the BRSs with which it is simultaneously connected, thus the terminal in macro-diversity uses more resources than those actually necessary for allowing the communications.
A further peculiarity of UMTS networks is that such networks are adapted to provide a plurality of different network services, such as, for example, telephony, fax, video-telephony, Internet access and Web browsing, streaming and so on. Each one of such services generally has characteristics in terms of speed (number of bits per second) and traffic (amount, symmetrical or asymmetrical) that are specific for the service under examination.
The dimensioning of the cells should therefore take into account both the characteristics of each service, and the possible associations of services over a single radio carrier, as provided for by the CDMA/WCDMA access technique.
Moreover, like every cellular radio-mobile system, also a UMTS network has common broadcast control channels in the whole cell area. Such channels convey system information, that are necessary for radio apparatuses (receivers) of the mobile communications terminals.
Due to the networks' peculiarities, the planning of UMTS networks is a complex task, requiring approaches that are substantially different from those used for previous cellular mobile telecommunications networks, particularly second-generation cellular networks like those complying with the Global System for Mobile Communication (GSM) standard, or with the Interim Standard (IS95).
In general, in view of a current network deployment, the planning process aims to produce, as results or outputs, the proper positioning of the BRSs in the geographic area under examination, and also allows determining the set of radio-electric cell parameters (e.g., antenna tilt, azimuth of the direction of maximum gain, radio power, etc.) and the allocation of the radio resources assigned to the network operator (for example, radio carriers). Such outputs are determined by the planning process in compliance with planning objectives, such as, for example:                granted extension of territory covered by the network service, within an area under planning;        maximization of the traffic to be managed among those provided within the area under planning.        
Various planning techniques for UMTS networks are known; according to the followed approach, these techniques can be grouped into two different classes: statistical planning techniques and deterministic planning techniques.
Statistical planning techniques are mainly based on an approach of the Montecarlo type (refer for example to the document 3GPP TR 25.942 v3.0.0 2001-06—“RF System Scenarios—Release 1999” specification). The term “Montecarlo simulation” usually denotes a static simulation composed of a set of statistically independent snapshots. After having fixed the scenario being studied, each snapshot consists in realizing a stochastic process generated starting from different distributions of users in the area being examined. At the end of every snapshot, network performance indicators are provided as results, and the procedure ends with the statistical analysis of various performance indicators provided by every snapshot. The number of snapshots shall be enough to guarantee statistical stability for the planning results. This methodology is rather specific, and it is particularly adapted for examining performances of a UMTS network of relatively limited geographic width; owing to its intrinsic slowness, due to the statistical convergence of results, this technique is not suitable for the analysis of UMTS networks intended to cover geographical areas comparable with those of an entire nation, such as, for example, Italy.
Though keeping the characteristic of a static analysis, the deterministic planning techniques systematically take into account all pixels of the territory on which the network will be planned. Differently from statistical methods, the deterministic methods exploit, as input data, a single users distribution, and a single simulation is carried out, without the need of a statistical aggregations of the results. Deterministic planning techniques are more suitable for planning UMTS networks intended to cover relatively large geographical areas, even if the planning result is generally less adherent to the evolving reality.
Irrespective of the approach followed, one of the phases of the methods for planning a cellular mobile telecommunications network of the type herein considered is the planning/evaluation of the network coverage in uplink, also referred to as “power control on the uplink”, i.e. the planning/evaluation of the coverage in the radio link from UEs located on the pixels of the area under planning to the BRSs. In this phase, the transmission power required per traffic channel to the UEs located on the pixels of the area under planning is calculated. For each pixel belonging to the area under planning, and for each network service, the cell is determined which requires the lowest transmission power to an hypothetic UE located on that pixel and using that network service: the cell thus determined represents the serving cell of that pixel, as far as that network service is considered. If the lowest transmission power required to the UEs calculated in this phase exceeds the maximum power deliverable by the generic UE (a parameter which is predetermined and forms one of the inputs to the planning process), the pixel under consideration is put in outage for excessive required power in uplink: in other words, given the current network configuration, a generic UE located on that pixel will not be in condition of using that network service, because the transmission power that would be necessary for doing this is too high. The set of pixels for which the generic cell is the serving cell in respect of the generic network service forms the “cell uplink service area” of that cell for that service in uplink. The set of cell uplink service areas for the various cells of the area under planning and for the various network services forms the “network global uplink area”. The set of pixels put in outage for excessive required power in uplink, for the generic network service, forms the “service outage area” in uplink.
Essentially, in the uplink power control phase, a plurality of sets of maps is determined, one set of maps for each network service; the generic set of maps includes in turn a number of maps equal to the number of cells of the area under planning: each map is formed by those pixels for which the respective cell (the serving cell) requires the lowest power in the uplink (there is a minimum threshold represented by the minimum UE transmission power), those pixels for which the required transmission power in uplink exceeds the predetermined maximum transmission power of the generic UE, for the generic network service, being eliminated.
In a following phase of the planning process, referred to as “power control on the downlink”, the planning/evaluation of the network coverage in the radio link from the BRSs to the UEs located on the pixels of the area under planning is carried out, so to ascertain whether the downlink is a limiting factor. For each cell of the area under planning, the transmission power per traffic channel that the generic cell should deliver is calculated, for each pixel belonging to the cell uplink service area of that cell and for each network service (i.e., for example, for the telephony, facsimile, video-telephony, Internet access, e.g., Web browsing, services). If the calculated power, for the generic pixel, exceeds the maximum power that the serving cell can deliver for a traffic channel in respect of the considered network service, that pixel is put “out-of-service” (“outage”) for insufficient power in the downlink. In other words, even if the generic UE located on that pixel would be able to deliver the necessary transmission power for communicating with the BRS, it is the BRS that is not in condition to sustain the necessary transmission power for providing that service. The set of pixels, belonging to the service area of the generic cell in respect of the generic network service, not being in outage, forms the overall service area of the cell in respect of that network service. The union of all the overall service areas for all the network services and for all the cells of the area under planning is referred to as the network global downlink area of the network (in the area under planning).
The downlink power control phase also encompasses a cell “capacity check” on the downlink: the overall power that, according to the above-mentioned calculations, is estimated to be required to the generic cell is compared to the maximum power that the (power amplifiers of the BRS of the) cell can deliver: if the calculated overall required power exceeds the maximum power that the cell can deliver, the cell does not pass the capacity check, and it might be necessary to modify the traffic distribution and/or the locations of the cells in the area under planning.
A UMTS network planning methodology is described in the document AC016/CSE/DR/P/091 entitled “STORMS Project Final Report”, 1997, developed under the STORMS (Software Tools for the Optimization of Resources in Mobile Systems) project, promoted by the European Union. The planning methodology described in that document provides for an analysis of the uplink capacity of the cells (BRSs) based on interference/noise limitations. The maximum cell capacity in uplink, in terms of maximum number of active calls per cell and per service type (CA, CB, . . . , CN) is determined by solving the following linear system:
      η    =                  (                  1          +                      f            extra                          )            ⁢                        ∑                      i            =            A                    N                ⁢                                  ⁢                              C            i                    ⁢                      SAF            i            UF                    ⁢          SIN          ⁢                                          ⁢                      R            i                                                  C        i                    C        REf              =          k      i                  i      =      A        ,    B    ,    …    ⁢                  ,    N  where η is a multi-service fractional load factor (with respect to the full load condition), assumed as the nominal load of the cell, fextra is a normalized inter-cell interference factor (which is characteristic of the environment), SAFiUL is a service activity factor in uplink for the generic network service, and SINRi is a target signal to interference plus noise ratio.
Factors ki describe the requested traffic mix in terms of ratio between the maximum active calls per each service and a reference one. These parameters have to be consistent with the correspondent traffic mix figures, which describe the offered traffic mix (in Erl) for the pixel under study. To derive these parameters, an iterative procedure is adopted. In fact, factors ki refer to the partitioning of the active calls among different services on the cell area, i.e., they correspond to the maximum number of circuits required to carry the offered traffic load on the cell area.
Thus, to evaluate them, the traffic load of the cell should be known. To derive it from the traffic load per pixel, which is the available input data, the cell dimension should be known. But this is in fact the final goal of the evaluation process. To solve this plight, an iterative process is proposed. A first, rough estimate of the cell area (i.e., number of pixels) is produced, and the correspondent traffic load per service is evaluated by multiplying the traffic value per service per pixel by the estimated number of pixels of the cell. Then, the traffic load is converted into the equivalent maximum number of active calls by means of the Erlang-B formula:Max_Number_of_Users=Erlang—B(Traffic_load; Loss_Probability).
The assumed loss probability is 0.01. Based on this estimate for the cell traffic load, factors ki are evaluated and substituted in the previous equation of η. A new value for the cell capacity is obtained by solving the equation. The process is re-iterated (using the equation solution as a new starting point for the estimate of ki) till it reaches the convergence. Finally, the correspondent maximum cell capacity (in Erl) is obtained by applying the Erlang-B formula to the final result of the previous step (by imposing a given loss probability).